Custom Made LD Module.

[Kenom]

how effecient do you think a LD Module made from aluminum would be? would have to be something exactly like the Aixiz module maybe a little larger in diameter and adding fins to it somewhere either around the diameter or whanot.

Why are they made of brass and not aluminum?

 

[Daedal]

Honestly... I very much agree with you on that one. Brass is a very malleable metal and is not very great at thermal conductivity. The only reason I see for using brass is that it allows for minor blemishes in the production process and would not damage the diode during the fitting process. (Just a wild guess)

Either way... What I know and live by is copper is your friend. It is the best electric conductor (not the clearest, but has the least resistance), and it is the best thermal conductor. Aluminum is nice because it's silver-colored, and a little more resistive to discoloration than Copper. I always use Aluminum in all my heat-sinking applications simply because I have tons of it readily available in all shapes and sizes. Copper would be my ideal choice, but I can't find blocks of copper for cheap... no way!

With that said... here is a nice table to summarize what I just said above

Aluminum
Thermal Conductivity: ~250
Electrical Resistivity: 0.0000000265

Brass
Thermal Conductance: ~109
Electrical Resistivity: 0.000000035

Copper
Thermal Conductance: ~400
Electrical Resistivity: 0.0000000168

Glass
Thermal Conductance: ~1
Electrical Resistivity: 10000000000 to 100000000000000

Gold
Thermal Conductance: ~310
Electrical Resistivity: 0.0000000221

Silver
Thermal Conductance: ~430
Electrical Resistivity: 0.0000000159

Steel
Thermal Conductance: ~45
Electrical Resistivity: 0.000000452

Thermal conductance is in units of (W/m K)
Electrical resistivity is in units of (Ohm meters)

So... To help answer your question... A module made out of Aluminum would be 2 and a half times as efficient as brass at dissipating the heat from the laser diode.

P.S. They don't call it Arctic Silver for nothing

Hope this explains it;
DDL

 

[Kenom]

I would tend to agree with you as far as copper being the medium of choice. I however have never seen a worked peice of copper like that. Maybe we can talk our buddy Tallaxo into making one for us to try or someone else that has access to a lathe. I at this point would be happy with Aluminum instead of brass.

 

[Daedal]

I never had access to a lathe around here. I'd love to get one... but let's see me get $60,000 ready to pop on one of these bad boys! As much as I love these things, I think they're overprice. :-[

Anyway. Copper is so amazing at conducting heat that when it's sitting out in room temperature and you're feeling hot. Gram a piece of copper and stick it to your face! The stuff is crazy! I do most my "cooling" work with computers an know for a fact that pure copper is like a natural freezer. I also know that it can hold itself together under extreme heat and can be shaped into any way you want. If you'd like, you can come up with the shape/size/dimensions of the heat-sink that you want to create and I can calculate the exact heat dissipation that it can diffuse. (Ya... I still have my books )

My personal preference is for a replacement endcap on the dorcy. If that were to be replaced with an all-copper block, add some circular fins, and make the adjustable collimator from AixiZ fit right in, and you have the perfect laser! At that point you can drive up the diode as far as you want. With enough fins and a tight fit for the AixiZ module in the block, the duty cycle has got to jump up significantly... if not even be efficient enough to eliminate it altogether.

GL and keep us posted;
DDL

 

[Kenom]

now I'm curious over at cpf bimmerboy indicated that heatsinking isn't going to be efficient for a laser diode. I'll quote him

"BTW, in most cases with a DVD laser, heatsinking won't be of much practical value. At rated drive levels, the mass of the module/lens holder alone is enough sinking for at least a couple or a few minutes of constant on... and you won't even feel the module get warm. And when overdriven, damage happens to the mirrored facets of the resonant cavity, which a heatsink cannot prevent. Thus the reason for short duty cycles. In this case again, you'll never even feel the module get warm, much less a surrounding heatsink."

I don't agree with that assessment. What do you think? Is the heatsink going to pull the prevent damage to the resonant cavity if it's rated up to 250mw pw

 

[Daedal]

Hmm... that is a very interesting way of looking at it. I'm going to have to read up on this a little and see what exactly he is talking about before I make my assessment. As far as I know, the damage to the lasing optics usually happens when the diode is overdriven. This diode was made to operate at 250mW and should be capable of doing it. My only concern is that the internals of the diode are not made to dissipate that much heat very quickly and thus we end up with an LED (broken optics). The other way a laser diode is damages is by supplying a huge amount of current/voltage that the micro-wire connecting the terminal with the optics just snaps because it has internal resistance (thus heat, thus it breaks).

I will have to look into this before I say for sure though;
DDL

 

[pseudonomen137]

Copper oxidizes very readily which can be detrimental to its conductive and electrical properties, and it is more expensive (there may be an ease of machining aspect in there too). Aluminum like like 6061 is cheaper, more available in diff shapes/sizes, relatively easy for a lot of types of machining, improves on the oxidization problem with copper, and does all this while still retaining a decent portion of the thermal conductivity of copper.

As for the diode, if you're running them CW at 250mW, you'll probably want extra heatsinking especially since you're overdriving it. CW at 80mW (I think that's the actual spec on most of them), you're probably talking around a quarter of a watt of heat produced at most, so the brass mass should be enough. In either case though, you'll want a bit of metal in addition to the can, but its not gunna take too much to heatsink it.

As for killing the diode, you have to watch out for BOTh too much heat and too much input power.
The end facets of the cavity in the diode are very efficient mirrors, but there will still be some absorbance. When there is too much optical power, or too much heat to these mirrors the slightest failure will increase absorbance at that point, and the entire mirror is ruined in a nearly-instantaneous chain reaction (referred to as COD - Catastropic optical damage). I think you can also kill the semiconductor junction, but COD is the main problem I hear about.

 

[Gazoo]

So what are the actual diode housings made out of? Brass or copper?

 

[timelord]

Brass I would think

 

[Daedal]

Nicely explained Pseudo. That is exactly what I was suspecting. I did not assume that the mirrors would absorb much heat, and whatever they absorb they should be capable of dissipating it. I do disagree with the amount of heat generated though.

Power in = Power out.

Power in = 2.5 V x .3 A = 0.75 W

Power out = 250mW (optical output) + heat

The heat dissipated at 2.5V is half a watt, at 5V drive it's about 1.25 W. That is pretty decent heat!

Gazoo, the module is made of brass from what I can see and tell. It is plated with a chrome paint for aesthetic reasons perhaps, but the inside seems to be Brass.

--DDL

 

[Gazoo]

I wasn't referring to the module. I mean the actual guts of the diode are mounted in a 5.6mm or 9mm can. What is the can made of? I think it is brass.

 

[pseudonomen137]

Nicely explained Pseudo. That is exactly what I was suspecting. I did not assume that the mirrors would absorb much heat, and whatever they absorb they should be capable of dissipating it. I do disagree with the amount of heat generated though.

Power in = Power out.

Power in = 2.5 V x .3 A = 0.75 W

Power out = 250mW (optical output) + heat

The heat dissipated at 2.5V is half a watt, at 5V drive it's about 1.25 W. That is pretty decent heat!

Gazoo, the module is made of brass from what I can see and tell. It is plated with a chrome paint for aesthetic reasons perhaps, but the inside seems to be Brass.

--DDL

Daedal, sorry for the confusion: I was referring to running it at 80mW output - which is technically the true CW spec on most of these diodes I think. 250mW is not the CW rating, but the peak power for quasi-CW type operation.

My thinking was >25% efficiency for the diode, so at the very most theres 320mW electrical input, less 80mW output, so ~0.24W max heat produced

If you're running them at 250mW, we know some of them can handle that during a pulse, but since you're overdriving you would probably want good heatsinking, or perhaps even some cooling so the diode doesn't die an early death.

As for the end facets (aka mirrors  ) in the laser cavity, when you consider a regular mirror, even at minimal efficiency (lets say 70%), if you shine a 250mW pen laser at it it'll probably dissipate enough heat itself that it won't phase the mirror at all.
However, in a laser cavity you're talking things on the proportions of micrometers, yet still the same 250mW+ in power (250mW is only the power that escapes the OC facet, there's more than that bouncing around in there and getting amplified). So we're talking an enormously high energy density, a miniscule mirror that won't be much of a heatsink, and only the slightest imperfection needed to start a chain reaction for the entire mirror to fail. Not a very ideal situation  


PS: Nice avatar, what is the violet/blue looking light in the center? (or were those edited in?)

 

[Daedal]

I wasn't referring to the module. I mean the actual guts of the diode are mounted in a 5.6mm or 9mm can. What is the can made of? I think it is brass.

Ya, probably brass... :-/

--DDL

 

[Daedal]

Agreed Pseudo... All are good and valid points. I apologize for misunderstanding you. And I hope the ones we get will be capable of dissipating the heat at a sufficient enough rate. I have contacted Sony to get some figures for the heat dissipation and facet efficiency. Should get a reply back sometime very soon I hope.

--DDL

 

[Daedal]

PS: Nice avatar, what is the violet/blue looking light in the center? (or were those edited in?)

Thanx Pseudo. This is the work of our local and wonderful Ragnarok ;D

--DDL

 

[Hemlock_Mike]

I use mostly brass and copper because I can solder it after machining parts to fit Aixiz modules. The pocket for pressing the LD into will work OK maybe once with aluminum but it depends on the alloy.
Any of these metals is fun to machine !!!

Mike